• Artificial Intelligence in Breast Cancer Care: Enhancing Diagnosis, Prognosis, and Personalized Treatment
  • Shakiba Sharif,1,* Asiyeh Jebelli,2
    1. Department of Cell and Molecular Biology, Faculty of Biological Sciences, Kharazmi University, Tehran, Iran
    2. Department of Cell and Molecular Biology, Faculty of Biological Sciences, Kharazmi University, Tehran, Iran


  • Introduction: Breast cancer (BC) is a significant cause of cancer-related mortality in women worldwide. The role of artificial intelligence (AI) in personal medicine for BC has been highlighted because of its accessibility to genomic and screening data. AI helps scientists to diagnose and give a prognosis of BC. This article focuses on AI functions in the diagnosis and prognosis of BC using machine learning (ML) in the case of deep learning and its subfields.
  • Methods: Specific genes have been recognized with a main role in raising and predicting tumors in breast tissue using different ML-based methods, including Recursive Feature Elimination with Cross-Validation-Logistic Regression (RFECV-LR), Recursive Feature Elimination with Cross-Validation-Support Vector Machine (RFECV-SVM), Random Forest (RF), Extra Trees, Least Absolute Shrinkage and Selection Operator (LASSO), and Genetic Algorithm (GA).
  • Results: Some immune genes have been discovered to be associated with tumorigenesis genes in breast tissue. A Disease-free survival (DFS) graph can also be drawn for each person using particular algorithms of AI: Random Survival Forest (RSF), MultiTask Logistic Regression and Cox proportional hazards model. The new personalized DFS for each patient could help the physician to choose the best treatment, and it also helps in prognosis knowledge. The conventional DFS was not that much sharp to predict the future of individuals in diseases like BC in that environmental and physical factors play a role in raising tumors. AI is not just useful in screening but also in interpreting and extracting information from breast images to detect malignant tumors. The best advantage of AI in breast imaging is decreasing recalls and unnecessary biopsies. Integrating the information obtained from imaging approaches with genomic, physical, and environmental information can create a method of personalized medicine for BC.
  • Conclusion: There could be some problems in using AI in different stages of cancer and patients with high risk or dense breasts. With all these limitations, AI has many impressive applications on different types of breast screening methods such as contrast-enhanced digital mammography, contrast-enhanced spectral mammography, ultrasound, digital breast tomosynthesis, and etc. AI knowledge with expanded subfields in many cases of information such as imaging, analyzing genomic data and extracting clinical data has helped developing personalized medicine specially about cancer treatment.
  • Keywords: Cancer, Artificial intelligence, Personalized medicine